The candidate will develop novel methods of applying bioinformatics to genome-wide association studies (GWASs) to discover novel genetic variants that influence risk for addiction-related diseases. GWASs have the potential to discover new biological mechanisms that cause disease. However, due to the large number of tests, statistical interpretation of GWAS data requires substantial resources such as large sample sizes and additional genotyping in independent replication samples. Even if a single nucleotide polymorphism (SNP) has a statistically significant association, linkage disequilibrium (LD) with other SNPs causes ambiguity that cannot be resolved statistically. It is impossible to test all gene-gene interactions in a GWAS because this is computationally and statistically infeasible. Where statistics ends, biology begins. The candidate will use the known biology of interactions between genes to first test the most biologically promising gene-gene interactions. Similarly, after a GWAS, the candidate will use his novel algorithms to systematically prioritize SNPs for further study, such as genotyping in a replication sample, by incorporating external bioinformatic databases. In summary, the candidate will study a priori biological hypotheses, and develop systematic methods of testing these hypotheses using GWAS data. Data will be utilized from multiple domains, including SNP/gene functional properties such as promoters and synonymy,transcription factor binding sites, evolutionary conserved regions, biochemical pathways, and gene expression. The amount and diversity of biomolecular annotation data from public databases is overwhelming. The candidate will develop methods designed for maximum viability and interpretability. The key validation of these methods is the discovery of novel genetic variants that influence disease, and the candidate will apply these methods to several case/control samples with extensive genotyping and addiction-related questionnaire data that are being developed at Washington University. The candidate will further his education in psychopharmacology and the biochemical basis of addiction, and also in molecular and computational biology. This will be done with a superb set of mentors, consultants, and coursework. Overall, this research has great potential to discover novel genetic variants and interactions that influence addiction-related diseases, and to tranform the candidate into a productive independent investigator. RELEVANCE (See instructions): This research aims to discover genetic variants and interactions that influence addiction-related diseases such as nicotine dependence. Understanding this complex genetic structure may lead to new methods of diagnosis and treatment for these deadly diseases. Tobacco-related disease, for example, is still the world's leadingscause of preventable death.